Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Improving Image Captioning with Better Use of Captions

About

Image captioning is a multimodal problem that has drawn extensive attention in both the natural language processing and computer vision community. In this paper, we present a novel image captioning architecture to better explore semantics available in captions and leverage that to enhance both image representation and caption generation. Our models first construct caption-guided visual relationship graphs that introduce beneficial inductive bias using weakly supervised multi-instance learning. The representation is then enhanced with neighbouring and contextual nodes with their textual and visual features. During generation, the model further incorporates visual relationships using multi-task learning for jointly predicting word and object/predicate tag sequences. We perform extensive experiments on the MSCOCO dataset, showing that the proposed framework significantly outperforms the baselines, resulting in the state-of-the-art performance under a wide range of evaluation metrics.

Zhan Shi, Xu Zhou, Xipeng Qiu, Xiaodan Zhu• 2020

Related benchmarks

TaskDatasetResultRank
Text-to-Image GenerationGenEval
GenEval Score67
360
Text-to-Image GenerationDPG-Bench
Overall Score83.5
265
Text-to-Image GenerationGenEval (test)
Two Obj. Acc87
221
Showing 3 of 3 rows

Other info

Follow for update